TOPLINE:

MRI-detected extramural venous invasion (mrEMVI) in patients with rectal cancer was associated with a more than fivefold increased risk for recurrence and a more than threefold increased risk for death; however, postoperative histopathologically detected EMVI (pEMVI) showed no significant prognostic value in a prospective single-centre cohort.

METHODOLOGY:

• Researchers conducted a prospective cohort study of 150 patients with biopsy-confirmed rectal adenocarcinoma, located within 15 cm of the anal verge, who were enrolled between October 28, 2013, and December 19, 2017, at a single institution in Norway.
• EMVI assessment was performed using high-resolution 1.5T MRI, independently reviewed by two consultant radiologists with 10 and 2 years of experience in gastrointestinal radiology who were blinded to prior reports and patient outcomes.
• Histopathologic examination of surgical specimens from 74 patients who underwent curative surgery without neoadjuvant treatment was conducted by a consultant pathologist with 10 years of experience.
• Primary outcomes included interobserver agreement for EMVI detection measured using the kappa coefficient and clinical correlations with tumour characteristics.
• Long-term survival outcomes including recurrence-free survival (RFS) and overall survival (OS) were analysed, with a median follow-up duration of 60 months.

TAKEAWAY:

• mrEMVI was identified in 52.7% of patients, with good interobserver agreement (kappa coefficient, 0.68), and pEMVI was confirmed in 45.9% of surgical cases, with only fair MRI-histopathology agreement (kappa coefficient, 0.40).
• A shorter mesorectal fascia distance remained an independent predictor of adverse outcomes for both RFS (adjusted hazard ratio [aHR], 0.92; 95% CI, 0.85-0.99; P = .03) and OS (aHR, 0.92; 95% CI, 0.86-0.99; P = .03).
• Patients with mrEMVI positivity demonstrated significantly poorer RFS (aHR, 5.22; 95% CI, 1.96-13.94; P < .001) and OS (aHR, 3.40; 95% CI, 1.34-8.61; P = .01) in the multivariate analysis.
• pEMVI showed no significant prognostic value for either RFS (P = .48) or OS (P = .95), and in the surgery-only subgroup, recurrence occurred in 23.5% of pEMVI-positive patients vs 17.5% of pEMVI-negative patients, with no significant difference (P = .57).

IN PRACTICE:

"Our results underscore that mrEMVI provides a robust and reproducible prediction of 5-year RFS and OS. As a reliable preoperative imaging biomarker strongly correlated with oncologic events, mrEMVI offers greater clinical utility than postoperative histopathological EMVI in informing therapeutic decision-making and potentially improving oncologic outcomes," the authors wrote.

"The absence of a survival association for pEMVI should be interpreted cautiously, as the smaller pathology subset and limited number of events indicate insufficient power rather than true prognostic irrelevance," they added.

SOURCE:

The study was led by Aida Kapic Lunder, Akershus University Hospital, Greater Oslo Region, Norway. It was published online on June 10, 2026, in European Radiology.

LIMITATIONS:

The study was limited by its single-institution design, the modest overall sample size, and the use of a 1.5T MRI scanner rather than a 3T system, which might have improved EMVI assessment, potentially underestimating the true prognostic strength of mrEMVI. A smaller pathology subset of 74 patients with a limited number of events indicated insufficient statistical power for pEMVI analysis rather than true prognostic irrelevance. The study was further limited by standard histopathologic constraints, including sampling variability, the anatomic complexity of the rectal vasculature, and the inability to trace vein courses in standard sections, which often led to underreported EMVI in pathology.

DISCLOSURES:

The study received funding from the South-Eastern Norway Regional Health Authority, Akershus University Hospital Research Foundation, and Akershus University Hospital. The authors reported having no conflicts of interest.

This article was created using several editorial tools, including AI, as part of the process. Human editors reviewed this content before publication.